Although current nosology defines schizophrenia (SCZ) and bipolar disorder as unique constructs with distinctive clinical characteristics, there exist significant symptomatic and familial overlap. Epidemiologic and genetic studies suggest that SCZ shares common susceptibility genes with psychotic bipolar disorder (PBD), as opposed to nonpsychotic bipolar disorder (NPBD), suggesting that they may also share similar structural abnormalities in gray and white matter in the brain. We hypothesize that: 1) PBD and NPBD will differ in measures of subcortical gray matter and white matter integrity, 2) abnormalities in PBD will be intermediate between SCZ and NPBD, and 3) there will be a significant relationship between lifetime psychosis severity and structural measures. The Aims of this proposal are to: 1) collect MRI scans from 40 PBD and 40 NPBD subjects and generate volume and shape metrics for the hippocampus, thalamus and basal ganglia, and measures of corpus callosum integrity, 2) compare these measures between PBD, NPBD, SCZ and controls, and 3) correlate lifetime psychopathology with structural abnormalities in PBD and SCZ. This 5-year program will involve training in specific neuroimaging methodologies and the neurobiology of bipolar disorder, as well as general research and ethics. Imaging methods used in this proposal will be Large-Deformation High-Dimensional Brain Mapping and Diffusion Tensor Imaging. Deanna Barch, Ph.D. (Dept. of Psychiatry, Washington University), John Csernansky, MD (Northwestern University), Joshua Shimony, MD, PhD and Abraham Snyder, PhD, MD (both at the Dept. of Radiology, Washington University) will co-mentor the principal investigator's scientific development. All co-mentors have substantial experience in various areas of neuroimaging and in the application of related research methods, as well as experience training postdoctoral fellows and NIH grant recipients. The proposal is also supported by a rich diversity of collaborators and consultants to provide education and guidance in various aspects of research. The principal investigator has completed residency training in Psychiatry and a Masters in Psychiatric Epidemiology (MPE) at Washington University School of Medicine in St. Louis. PUBLIC HEALTH RELEVANCE: In clinical practice, treatments used in schizophrenia and bipolar disorder often overlap, and selected medications are not effective in all those affected by these disorders. Identification of specific structural brain abnormalities that cross traditional diagnostic boundaries, as is proposed, would allow for the development of more specific treatments for those affected, and identify those at risk. Specific biological abnormalities will also help validate psychiatric diagnoses and improve the classification of disorders.